slow, dumb, cheap compound setup!

Mike_04GolfTDI

Top Post Dawg
Joined
Nov 19, 2003
Location
Richmond, BC, Canada
TDI
Mine: 2019 Golf R DSG, Wife's: 2015 Golf Comfortline TDI
Once I made a dual alternator setup for a Chevy Blazer that I had, because the one alternator wasn't putting out enough power to run the powerful stereo that I had in it.

With two alternators it worked great.

A while later, I came to realize that the original alternator just didn't work properly and only the 2nd one was really doing anything.

http://www.youtube.com/watch?v=1ytCEuuW2_A
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Oh, the stock turbo is still doing most of the work and all.
I'm just fixing it until it's broken.
My MAF values go up to 1200mg/st of air (EGR off) with 17psi of boost, so I changed the X axis values in the 13x16 map to: 850, 890, 930, 1100. You SHOULD raise the X axis values for better control because you have much more airflow that needs to be defined.
That is the info I needed to know.

Every scan tool I plug in freezes as soon as the engine is running.

How far is the stock MAF good for?
 

Enabled

Veteran Member
Joined
Apr 23, 2013
Location
Houston, TX
TDI
2003 Jetta TDI Manual, BMW 328d SW
My stock MAF has hit a ceiling at 1270mg/stroke of air. This is calculated by the ECU or VCDS. The MAF reports in voltage based on air mass per time. We'd have to calculate it out to find out in the usual flow units, g/sec, Kg/Hour, etc.

I think that's the maximum voltage it puts out, so I might be getting even more air than 1270mg/str, but I'm happy I'm metering that much, for 50mg (plus whatever extra my PP520's give out, probably 5-10mg of extra fuel).

Also you shouldn't have to change Start of Injection, or Pump Voltage map (AKA Duration Map, at hex address 54470 manual, 74470 auto codeblock) too much, as they're both defined in the X-axis up to 51mg of fueling from factory. With nozzles putting out a little more uncalculated fuel, you may need to then.
 
Last edited:

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Also you shouldn't have to change Start of Injection,
I've advanced the **** outta my timing already, mainly to get the fuel efficiency up to 50ish from a dismal 42.

Still plenty of room to go further, all my changes to it have resulted in better economy, but I'm not certain the thing'll handle too much more. On my truck you notice the temp gauge coming up faster and you shut it down, loosen three bolts and bump the IP back a hair. The bottom end is built strong (like tractor :p )
I'm not even sure this VW's coolant temp gauge works, it slowly comes up to 190 then just stays there no matter what. No thermostat works that well, it should fluctuate at least a tiny bit.
Pretty sure it's just a slightly smarter version of the oil pressure idiot-light "gauges" found in every GM and ford product from 1992 on.
 

Enabled

Veteran Member
Joined
Apr 23, 2013
Location
Houston, TX
TDI
2003 Jetta TDI Manual, BMW 328d SW
You advanced it with Start of Injection maps, or with actual on pump adjustments (and comparing the timing graph, mines on the top line), or VCDS basic settings?

I advanced mine at the pump, and still haven't noticed better gas mileage. I can hit 47-50 if I drive normally and no more than 60mph.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
You advanced it with Start of Injection maps, or with actual on pump adjustments (and comparing the timing graph, mines on the top line), or VCDS basic settings?

I advanced mine at the pump, and still haven't noticed better gas mileage. I can hit 47-50 if I drive normally and no more than 60mph.
Advancing it at the pump does nothing but at startup.

If you advance it at the pump to the point of throwing a code (WAY off the timing graph on the vagcom) you're still a few degrees behind even stock timing at 2k RPM.

Gotta do it in the maps to see benefit above idle. I'd post a screenie of my maps, but I don't have this linux thing dicked quite yet. I more or less copied this guy's map a while back, though.
http://forums.tdiclub.com/showpost.php?p=4526669&postcount=281

I get 49 driving like a dick.

Well, reflashing the thing for the 11th time tomorrow. Higher upper limit on the smoke map (up to 1100 mg/str of air up from 853) and some fault codes disabled relating to the EGR valve solenoid and the GP controller.
 

Enabled

Veteran Member
Joined
Apr 23, 2013
Location
Houston, TX
TDI
2003 Jetta TDI Manual, BMW 328d SW
Advancing it at the pump does nothing but at startup.

If you advance it at the pump to the point of throwing a code (WAY off the timing graph on the vagcom) you're still a few degrees behind even stock timing at 2k RPM.

Gotta do it in the maps to see benefit above idle. I'd post a screenie of my maps, but I don't have this linux thing dicked quite yet. I more or less copied this guy's map a while back, though.
http://forums.tdiclub.com/showpost.php?p=4526669&postcount=281

I get 49 driving like a dick.

Well, reflashing the thing for the 11th time tomorrow. Higher upper limit on the smoke map (up to 1100 mg/str of air up from 853) and some fault codes disabled relating to the EGR valve solenoid and the GP controller.

Thanks for the link. I'm really looking forward to see if it really makes a difference. All the other VCDS changes .. and hammering .. and grease on the MAF connectors .. and advance from the pump .. and advance from VCDS basic settings did nothing for my mpg.

I'll make start of injection that way. Only to the 86 degree map I suppose. I could use some mpg back. I've flashed mine about 15 times by now. Glad this MPPS cable is working well without flaw. Very happy with the power, finally it's acceptable for me to get around.

Working on my 26th version right now. V1.52..
 
Last edited:

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Only to the 86 degree map I suppose.
I went with both 86 and the 81 (only changing the values where the 81 ended up with less advance) to begin with and it did make quite a bit of difference.
Recently went through the maps back in temperature to bring the other maps up to at least the baseline of that 86 degree map.

Timing should come down some as it heats up, so really, the cold maps should be full of numbers slightly higher than the warmer ones, as they are in the factory maps.
 

Enabled

Veteran Member
Joined
Apr 23, 2013
Location
Houston, TX
TDI
2003 Jetta TDI Manual, BMW 328d SW
I went with both 86 and the 81 (only changing the values where the 81 ended up with less advance) to begin with and it did make quite a bit of difference.
Recently went through the maps back in temperature to bring the other maps up to at least the baseline of that 86 degree map.

Timing should come down some as it heats up, so really, the cold maps should be full of numbers slightly higher than the warmer ones, as they are in the factory maps.
I had copied my new 86 degree map also to the 81 degree map right after finishing the 86, because the 81 didn't "fit" anymore with all that after top dead center.

I left the bump at low RPM and high fueling the same though at around 11 BTDC, as I'm happy with my starting performance at hot and cold engine.
I'm really looking forward to better mpg. Hopefully EGT's won't be too high.

Here it is, if not too intrusive to your Compound turbo thread. If it is, I'll remove it.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
I had copied my new 86 degree map also to the 81 degree map right after finishing the 86, because the 81 didn't "fit" anymore with all that after top dead center.

I left the bump at low RPM and high fueling the same though at around 11 BTDC, as I'm happy with my starting performance at hot and cold engine.
I'm really looking forward to better mpg. Hopefully EGT's won't be too high.
Yup, lower right is for the start map, wouldn't get that kind of fuel at that low of RPM through any other avenue. Doesn't really do much though, other than in an extended crank in very cold weather. Pump advance is controlled through internal fuel pressure from the transfer pump. Computer switches a solenoid that restricts the fuel return from the pump and that actuates an advance piston. Engine's usually running before you get pressure in the pump case and therefore computerized advance, runs for that second on that 1/2qt reservoir of fuel in the pump.
EGTs should go down quite a bit, as the fuel is being burnt in the combustion chamber rather than in the exhaust (to keep the catalyst hot).

No issues, learning's going on. I was hugely frustrated with the lack of info not even a year ago.

ETA: as an aside, got the GP codes disabled and the "relay" out and halfway converted to pushbutton control. Just gotta do the work thing for a bit then I'll get it done.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Stuck a drum brake shoe return spring on the thing and now it makes 10 PSI. It's dark out and my gauges are not exactly lighted, so driving hard with the dome light, headlights and the wipers going is a little dangerous. :p

So far no limp mode or anything.
VNT seems to still be able to keep the boost under control for the short little 2k-3.5k RPM pull that I was watching the IMP gauge for.
 

andy2

Veteran Member
Joined
Sep 24, 2004
Location
Bowmanville,Ontario,Canada
TDI
13 Jetta,94 Golf drag car 585bhp,Samurai buggy BHW 300bhp,97 Ram cummins
As long as your n75 duty cycle can keep MAP within the limit without setting an overboost code then you are ok.When you get the vanes on your high pressure turbo to open all the way and still not overboost on the MAP then EMP's should be at its lowest and most efficient.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
As long as your n75 duty cycle can keep MAP within the limit without setting an overboost code then you are ok.When you get the vanes on your high pressure turbo to open all the way and still not overboost on the MAP then EMP's should be at its lowest and most efficient.
That's what I'm figuring, so far fuel economy's down though.
Probably because now I've got three gauges to watch bounce around and watch for patterns in.

Made up another remap. More fuel this time. DW map was limiting it this time. Guess 100% throttle doesn't really mean 100% at anything above 2k RPM or something. Fixed that.
Might flash it later tonight. Probably not, gotta get to school with the car, so probably do it tomorrow around noon, as then I'll 'only' have to get to work. heh

ETA: Really gotta get around to putting together the smaller fuel tank. Less weight, less mileage between fillups (easier to do a drive like a granny economy tank for calculations).
 
Last edited:

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Okay, got the motor out and on the stand. Got some ARLs in standard size (by mistake) and some chinese rods. Some of these for the main bolts https://www.fastenal.com/web/products/details/1139627 and I'll be re-using my ARP head studs. They were in there tight as all hell, 125 ft/lb after a few heat cycles, I had to rattle on them with the IR231 for a couple seconds apiece to get them to come loose.

There's still crosshatch on the cylinder walls, no scratches or anything. I hear factory piston/wall clearances are 1 to 2 thousandths, guys with hot B-series motors run .009" or so clearance for the pistons to heat up and expand. These pistons are 3/4 the size, so .006" should be good, I'm gonna try for .005" clearance from honing the cylinders.

Chinese rods are getting the big ends honed out proper. Then I'll assemble it with the old bearings to measure my piston protrusion and figure out a head gasket from there. Had a 2 hole, but with new pistons and rods, who knows where it'll end up.

So I don't forget the numbers, I'll transcribe them from the pay stubs I have them written on.

ring end gaps
1 0.5mm
2 0.4mm
oil 0.3mm
rod bolts, 46 ft/lb
mains, 50 ft/lb

ETA: had about 1800 miles on the compound setup, still on a 2.5bar map, not much to say, the subaru turbo'd make 10 after adding a drum brake spring to the wastegate, and then the rest of the 21 or so would get done by the VNT15

ETA: 12/7/14


Lotsa metal came outta the head. Exhaust ports were TINY. Maybe to keep velocity up so soot couldn't build up in the head where it is cooled by the water jacket? Dunno, hogged them out to a little smaller than the gasket, at least 1/8" in every direction. Smoothed out the manifold a little too, but didn't remove very much at all.

Popped some new guides in and reamed them, gotta get together a setup to grind the seats, Mandrels and everything's dirt easy, there's a bin of nice new pink valve grinding stones at the surplus shop, and I can dress them to the proper angle on the lathe with sacrificial carbide tooling.

Honed the cylinders earlier today, did the first two good to .0001 taper, then on the last two got tired and said .0004 was good enough. Should have .004 or so piston-wall clearance. Little tighter than I was shooting for, but I'm not going too nuts with fueling, and they shouldn't get too hot with the oil passages in them...

So, tomorrow, new pistons and rods go in to figure out what head gasket to order, then back apart for some paint and new con-rod bearings when they get here.

ETA3: 12/28/15
Accidentally sunk a couple intakes far too deep, can't find ANY valve seats that'll work from the (read: any) catalog. Intake valves are commonly made outta 4140, I was going to use 4130 for the seat, but I had some 1045 laying about and it'll be nicer to the $90 cutter blades on the valve seat cutter. Made it a .008 press fit, tossed the head in the oven at 275 and the seat in the freezer, hammered them together with a 3lb drilling hammer. Really would have rather used something closer to 1lb, for the sharpness of blows rather than the "move metal and bend the head" that the 3lb hammer offers against the cam cap studs it was siting on.

Later on when I find a junk head to figure out a compression release brake on, I'll make some shrink fit steel inserts that go out to the edge of the fire ring portion of the head gasket. Should fix any propensity to cracking between the exhaust seat and the injector hole. Then I'll go to 1.5" OD intake seats, and whatever fits for exhausts. Maybe some 5.5mm honda valve stems, too. Maybe roller rockers from a mitsu v6 like you find in older caravans. Probably just do lash cap adjustable solid buckets with individual hydraulic passages to the exhausts. Separate little hydraulic cylinder mounted at the right spot to catch a cam lobe and open the valve a couple hundredths, something like 10 degrees before TDC. Solenoid fills the loop with oil, leakage through the lifter bores lets the system drain down when off the 'jakes'

eta: 1-10-15, runs and drives, screwed up the assembly of the VNT mech, so it only can clamp the vanes down about halfway before it hits the spacers (I put them around the screws in between the vanes rather than around the pins, oops) So I'll be pulling it down and fixing that tomorrow, checking the cam timing and probably backing off the IP timing a little tiny bit, got it advanced to a minimum of 7.5 degrees BTDC, was hoping for around 6 degrees minimum advance. Might even get some pics of the compounds in their natural habitat, from the underside to boot. Got a 4 BAR map ordered, so I'll be able to actually tune for more pressure.
Plans are in my brain for a water/air intercooler integrated into the intake manifold, made out of an old A/C evaporator that had a pinhole leak in the piping. Right now I'm using the stock intake system from the VNT turbo to the intake ports in the head. Plastic tanks usually don't live long beyond 20 PSI.

Just keeping my thoughts ordered (as they get)
 
Last edited:

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
I really suck hard at getting decent pics. Gotta get better at that, anyways, engine rebuild is done, I edited in a bunch of details to the previous post ITT in the intervening months. Read through it if you feel like it. Ported the head and stuff.

Looking at it from the top, you can sorta kinda see the exhaust manifold on the bottom, and my ghetto brass manifold for EGT/EMP readings.


From the underside, if you squint you can see the aluminum oil pan and trans case, the CV axle and the swaybar.
 

fouckhest

Veteran Member
Joined
Aug 26, 2007
Location
g'vegs, sc
TDI
.:GTD, .:R32, DangerRanger
Want to buy a proper manifold and up pipe setup and take your compound setup to the next level?
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Want to buy a proper manifold and up pipe setup and take your compound setup to the next level?
Eh, not really. I'd be interested in a stock ALH intake manifold (or really anything that bolts up to an ALH head) though.

Got the idea in my head to weld in a section of A/C evaporator core as a water/air intercooler. It'd also probably screen out smaller blown up turbo bits reasonably well. :p

Run a coolant line from the water outlet on the head (probably get the plastic adapter in the middle of the upper rad hose from a 1.8T or an automatic, those have an extra fitting on them) to a repurposed A/C condenser out front to get it down to ambient temperature. Then run it through the heater core and dump it into the heater core return. Then I can avoid running a separate pump/reservoir/relay, and still have relatively cool coolant running through the intercooler. Less piping, more room to get my hands in to work on it.

If you mean the exhaust side, nah, I'll let the VNT-15 run its course, then I'll make a square tube header for it like I did for my datsun. Always good for a lot of laughs when people see it. Maybe run a k03 that's sitting on the shelf, that's a bit bigger than the VNT 1749vc, even.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Don't think you'll flow much water through a/c cores.
That's the point, then it doesn't put the coolant temp too low. It'd be a loop of the engine cooling system with no thermostatic control.

I figure that water has such a comparatively huge thermal mass it doesn't need quite so much flow through it to keep the small intake charge cool.

The point is to avoid a separate pump. The evaporator core is relatively high flow,(obviously once the metering device is removed) and the condenser can be selected appropriately. As an aside, they make great A/T coolers for those stuck with an auto trans in a truck.
 

Digital Corpus

Top Post Dawg
Joined
Mar 14, 2008
Location
Ontario, California
TDI
'97 B4 w/ 236K mi body, 46K mi soul
Water has ~4x the heat capacity than air. Don't know what it is of the oil. However, despite the higher capacity, which means it'll absorb a lot more heat, it'll take that more extra work to release it back to the air.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Water has ~4x the heat capacity than air. Don't know what it is of the oil. However, despite the higher capacity, which means it'll absorb a lot more heat, it'll take that more extra work to release it back to the air.
What I was explaining is that since the water is denser it doesn't need a huge amount of flow. A 3/4" line like to the heater core would be very much excessive. A 1/2" line should be plenty, especially with a simple little water injection setup after the cooler.

It being harder to release into the air is a non-issue, with the 4+ square feet of frontal area most condensers take up, there's plenty of room to bring it down around ambient temp.
 
Last edited:

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
So, I've got a 4bar map that is close to properly linearized, I've pegged my 20 PSI boost gauge back and around to the vacuum side. I'm going to estimate something like 35 PSI. Oops. I only really intended for something rather mundane like 25 or so, but hey, the VNT-15 actually was able to flow enough to pressurize the intake manifold that much. Surprised the hell out of me after having looked into the compressor outlet and seen what looked like the opening in the end of a garden hose.

The spring piggybacked onto the gate of the subaru turbo's gotten weaker, probably from heat, and it only makes 7lbs or so. Going to have to get some more spring to it. Opinions on welding the 'gate shut? I don't think I'm likely to damage that thing with a cute little 1.9L diesel.


ETA: Finally did some datalogging, or at least looking at the vagcom while driving. Figured out what makes it wig out and stop working, when I use the starter it screws up the communication, so I just push start it and it's happy.
Getting closer to proper linearization, now 960mbar atmo is reading as 999mbar, and 17.5ish PSI is reading as 2500 absolute. Working on it some more.
I can easily max out the 51mg/str fueling through the very conservative 17:1 smoke map, I think I saw something around 1000mg/str of airflow at that 17.5 PSI of boost, so when I get the bigger injectors in there I'll have to look about for a 3" MAF housing.

I was able to get the N75 map tweaked so the boost spikes are minimal. More duty cycle means less vacuum on the actuator, and therefore more open vanes. Changed the values from sixty somethings in the upper right of the map to 75, seems to have done the trick. Response is a little bit slower, but I don't mind that quite so much. Only changed the rightmost column, so anything below 45mg/str or whatever of pedal is unaffected.
 
Last edited:

Alcaid

Veteran Member
Joined
Sep 5, 2006
Location
Norway
TDI
See signature
Do not weld the gate shut! Never do that in a compound unless the LP turbo has a very big turbine, your's don't.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Do not weld the gate shut! Never do that in a compound unless the LP turbo has a very big turbine, your's don't.
I was thinking that the increased pressure across the LP turbine would help with the choke flow across the itty bitty HP turbine. Bandaid fix to the real problem. Thanks for the slap to the wrist to bring me back to reality. :p

I think I'll leave it alone for now, as it is very well able to burn all the fuel I can inject at the moment. I've got larger injectors and a 12mm pumphead on the shelf though, so I am looking for the direction to grow.

I've got a K03-005 on the shelf from a 1.8T, and the compressor looks to flow an awful lot more than the gt1749v, but I can't find a compressor map to compare the pressure ratios it'll support. Turbine housing a/r looks tiny, but it might be able to be stuck onto the 1749v's turbine housing. I've got a lathe, mill, and the audacity to try, provided the compressor would support the pressure.

ETA:Lost to a mini cooper S going 0-60mph from a stoplight. hahahah
I heard a turbo that wasn't mine, quick wikipedia search says he had 175hp from the factory, and about the same weight.
 
Last edited:

dave926

Member
Joined
Jul 3, 2013
Location
Barrington, RI
TDI
Jetta 1.8t
So, I've got a 4bar map that is close to properly linearized, I've pegged my 20 PSI boost gauge back and around to the vacuum side. I'm going to estimate something like 35 PSI. Oops. I only really intended for something rather mundane like 25 or so, but hey, the VNT-15 actually was able to flow enough to pressurize the intake manifold that much. Surprised the hell out of me after having looked into the compressor outlet and seen what looked like the opening in the end of a garden hose.

The spring piggybacked onto the gate of the subaru turbo's gotten weaker, probably from heat, and it only makes 7lbs or so. Going to have to get some more spring to it. Opinions on welding the 'gate shut? I don't think I'm likely to damage that thing with a cute little 1.9L diesel.


ETA: Finally did some datalogging, or at least looking at the vagcom while driving. Figured out what makes it wig out and stop working, when I use the starter it screws up the communication, so I just push start it and it's happy.
Getting closer to proper linearization, now 960mbar atmo is reading as 999mbar, and 17.5ish PSI is reading as 2500 absolute. Working on it some more.
I can easily max out the 51mg/str fueling through the very conservative 17:1 smoke map, I think I saw something around 1000mg/str of airflow at that 17.5 PSI of boost, so when I get the bigger injectors in there I'll have to look about for a 3" MAF housing.

I was able to get the N75 map tweaked so the boost spikes are minimal. More duty cycle means less vacuum on the actuator, and therefore more open vanes. Changed the values from sixty somethings in the upper right of the map to 75, seems to have done the trick. Response is a little bit slower, but I don't mind that quite so much. Only changed the rightmost column, so anything below 45mg/str or whatever of pedal is unaffected.
If you want to send me your file Ill try to find your MAF/MAP switch and the MAP based IQ limiter
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
If you want to send me your file Ill try to find your MAF/MAP switch and the MAP based IQ limiter
Hm? MAP based smokemaps may be in the future, but now I've realized I've only got enough fuel for about 900 mg/stroke of air. Lots of headroom.
Another realization is that while my engine used to smoke a little tiny bit at 17:1 or so on the smoke maps, now that I've hogged out the exhaust ports it is clean burning even after tweaking the smoke maps to 16.5:1, I'm going to go further to 16:1 next.

Going to be working on my PP764 injectors a bit more tomorrow, realized that my method of setting the pop pressures has changed the maximum pintle lift. Bit of a brain fart there, oh well, some measuring and fitting and she'll be right as rain.
 

[486]

Top Post Dawg
Joined
Mar 1, 2014
Location
MN
TDI
02 golf ALH
Arright, got the injectors in there and oh man it smokes when you get on it, even after raising the smoke maps up to 18:1, gonna go to 20:1 next. I'm going to go that way rather than screwing with the pump voltage map, because then I don't have to screw around with getting the computer to put out better than 51mg/str. I'll just scale the smoke maps and all that jazz. MAP sensor is getting closer to proper linearization as well.

Oh right, I set up the injectors as single stage, with about .004" more pintle lift than stock. Got the pintle lift to within .0002" across all four, and the pop pressures are all within 50 PSI, no idea where they are (gauge went off the scale, so I moved the needle back onto the scale), but they are within 50 PSI of each other. hah

Isn't much louder at idle or otherwise, was hoping for more rattle, but hey guess that means that the PCPs are nice and low, more timing is in order once I get the MAP properly calibrated...

ETA: It spins the tires in second just mashing the throttle, no clutch abuse or anything. Hah.
 
Last edited:
Top